Research map / 3D vision / NeRF and radiance fields
GRF: Learning a General Radiance Field for 3D Representation and Rendering
Alex Trevithick, Bo Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add input viewpoints to the CNN module (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | PSNR: 20.13 -> 27.16 | Improved |
| add input viewpoints to the CNN module (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | SSIM: 0.807 -> 0.942 | Improved |
| add pixel local features (CNN decoder) instead of a global feature vector (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | PSNR: 20.23 -> 27.16 | Improved |
| add pixel local features instead of a global feature vector (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | SSIM: 0.818 -> 0.942 | Improved |
| use attention aggregation instead of max-pooling (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | PSNR: 24.88 -> 27.16 | Improved |
| use attention aggregation instead of max-pooling (ablation) | GRF · ShapeNetv2 Cars (ablation subset) | SSIM: 0.914 -> 0.942 | Improved |
| increase input views from 2 to 6 | GRF · ShapeNet unseen 7 classes | PSNR: 24.68 -> 29.37 | Improved |
| increase input views from 2 to 6 for SRNs (retrained) | SRNs · ShapeNet unseen 7 classes | PSNR: 24.16 -> 25.76 | Improved |
| finetune pretrained GRF for 100 iterations instead of training NeRF from scratch | GRF · Synthetic-NeRF (4 novel scenes) | PSNR: 15.15 -> 19.69 | Improved |
| finetune pretrained GRF for 10k iterations instead of training NeRF from scratch | GRF · Synthetic-NeRF (4 novel scenes) | PSNR: 23.35 -> 25.10 | Improved |
| finetune pretrained GRF for 1k iterations vs none | GRF · Synthetic-NeRF (4 novel scenes) | LPIPS: 0.246 -> 0.128 | Improved |
| use GRF instead of NeRF for single-scene learning | GRF vs NeRF · Real-world scenes (8 scenes) | LPIPS: 0.250 -> 0.178 | Improved |
Closest papers by meaning
- Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenes (CVPR 2021 · similarity 0.91)
- Learning Robust Generalizable Radiance Field with Visibility and Feature Augmented Point Representation (ICLR 2024 · similarity 0.91)
- GSNeRF: Generalizable Semantic Neural Radiance Fields with Enhanced 3D Scene Understanding (CVPR 2024 · similarity 0.90)
- Mask-Based Modeling for Neural Radiance Fields (ICLR 2024 · similarity 0.90)
- HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis (NeurIPS 2025 · similarity 0.89)
- GSN: Generalisable Segmentation in Neural Radiance Field (AAAI 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
